US20100287413A1

US20100287413A1 - Error detection and/or correction through coordinated computations

Info

Publication number: US20100287413A1
Application number: US12/463,979
Authority: US
Inventors: Miodrag Potkonjak
Original assignee: Individual
Current assignee: ARISTAEUS HERMES LLC; Empire Technology Development LLC
Priority date: 2009-05-11
Filing date: 2009-05-11
Publication date: 2010-11-11
Also published as: US8566638B2; US20120311382A1; US8255743B2

Abstract

Techniques are generally described for addressing computation errors via coordinated computation on two computing platforms are disclosed. In some embodiments, one or more cuts may be taken of a computation to observe variables, and the observations may be analyzed to detect errors. Corrections may be created for the detected errors. The disclosed techniques may be employed in power and/or energy minimization/reduction, and debugging, among other goals. Other embodiments may be disclosed and/or claimed.

Description

BACKGROUND

There may be numerous potential sources of errors in computational, communicating, and data storing systems and data processing including algorithmic and data structure errors, compilation mistakes, errors in design specification and design implementation, manufacturing faults, environmental impact, inadequate operational conditions, data alternations, and intentional malicious attacks. Essentially all technological and application trends may be likely to have additional negative impact on the error rates and their impact. For example, feature scaling may exponentially increase the likelihood of radiation errors. Also, the exponentially growing rate of integration may make difficult to produce manufacturing error free integrated circuits and systems. Finally, the complexity of application grows at significantly higher rates than the computational capabilities of silicon. One of the ramifications of a more efficient use of silicon may be increased number of design errors. Even more importantly may be that optimization of all design metrics, including clocking speed, energy, area, manufacturing and testing cost, latency, throughput and reliability, may be inevitably related to the ability to address error. And the ability to address error may be particularly important to low power and/or energy minimization/reduction, and debugging, among other goals.

BRIEF DESCRIPTION OF THE DRAWINGS

Subject matter is particularly pointed out and distinctly claimed in the concluding portion of the specification. The foregoing and other features of the present disclosure will become more fully apparent from the following description and appended claims, taken in conjunction with the accompanying drawings. Understanding that these drawings depict only several embodiments in accordance with the disclosure and are, therefore, not to be considered limiting of its scope, the disclosure will be described with additional specificity and detail through use of the accompanying drawings, in which:

FIG. 1 illustrates an overview of computation error detection and/or correction through coordinated computation, in accordance with various examples;

FIG. 2 illustrates a method to address computing errors, in accordance with various examples;

FIGS. 3 a-3 c illustrate an example computation cut and variable observation, in accordance with various examples;

FIG. 4 illustrates an example processor with implementation decisions that facilitate addressing errors, including error detection and correction through coordinated computation, in accordance with various examples;

FIG. 5 illustrates a computing system configured in accordance with various examples; and

FIG. 6 illustrates a computing program product in accordance with various examples, all arranged in accordance with the present disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS

The following description sets forth various examples along with specific details to provide a thorough understanding of claimed subject matter. It will be understood by those skilled in the art, however, that claimed subject matter may be practiced without some or more of the specific details disclosed herein. Further, in some circumstances, well-known methods, procedures, systems, components and/or circuits have not been described in detail in order to avoid unnecessarily obscuring claimed subject matter. In the following detailed description, reference is made to the accompanying drawings, which form a part hereof. In the drawings, similar symbols typically identify similar components, unless context dictates otherwise. The illustrative embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be utilized, and other changes may be made, without departing from the spirit or scope of the subject matter presented here. It will be readily understood that the aspects of the present disclosure, as generally described herein, and illustrated in the Figures, can be arranged, substituted, combined, and designed in a wide variety of different configurations, all of which are explicitly contemplated and make part of this disclosure.
This disclosure is drawn, inter alia, to methods, apparatus, and systems related to addressing computation errors via coordinated computation on two computing platforms. Embodiments of the present disclosure include various techniques to address the issue of computation errors through coordinated computation. In various embodiments, a technique for addressing computation errors comprises methods for error detection, diagnosis, characterization, and correction. In various embodiments, another technique includes synthesis and implementation techniques for facilitating errors addressing. In various embodiments, the techniques may be employed to correct permanent errors due to manufacturing variability.
In the following detailed description, reference is made to the accompanying drawings which form a part hereof, and in which are shown by way of illustration embodiments in which embodiments may be practiced. It is to be understood that other embodiments may be utilized and structural or logical changes may be made without departing from the scope of the disclosure. Therefore, the following detailed description is not to be taken in a limiting sense, and the scope of embodiments is defined by the appended claims and their equivalents.
Referring now to FIG. 1, wherein an overview of the present disclosure of addressing computing errors through coordinated computation, in accordance with various embodiments, is illustrated. As shown, for the embodiments, two different processors or computing platforms 101 and 103 may be configured to coordinate execution of the same or similar functionality. At least one of these executions, e.g. execution on processor or computing platform 103, may be conducted in such a way that it may be significantly more robust against large set of errors. Therefore, this or these types of execution may be used as a standard for correcting errors in the other execution, e.g. execution on processor or computing platform 101, done under energy constraints or any other setting where the errors may be more likely due to reasons such as radiation or high temperature.
In various embodiments, processors or computer systems 101 and 103 may be independently provided with the program and input data to be executed in a coordinated manner. At action 105 (Cut Probe), in various embodiments, processor or computing platform 103 may be configured to instruct processor or computing platform 103 on where and when to take a cut of the computation and observe one or more variables at the cut, to be described more fully below. At action 107 (Observed Variables), processor or computing platform 101 may be configured to send real time variables that may be used for error detection to processor or computing platform 103 that executes the same or similar program using the same or similar input data. In various embodiments, on receipt, processor or computing platform 103 may analyze the received variable values, and detect for errors based at least in part on the result of the analysis. At action 109 (Corrections), in various embodiments, processor or computing platform 103 further may diagnose the errors, create corrections to correct one or more of the detected errors, and/or may send instructions to processor or computing platform 103 on how to correct the detected errors.
In alternate embodiments, the processor or computing platform 103 may characterize processor or computing platform 101 in terms of its faults, simulates or emulates its execution and sends the corrective instructions to processor or computing platform 101. The correction may be either data dependent or generic for a pertinent program.
In various embodiments, the present disclosure may be practiced in an off-line manner; in other embodiments, the present disclosure may be practiced in an on-line manner. In an on-line embodiment, the data may be transferred between the involved devices using either wired or wireless communication. In various embodiments, one of the executions may be conducted using tight clock cycle time of a fixed point processor and the other may be done with lax clock cycle time on a floating point processor.
Before further describing embodiments of the present disclosure, it should be noted that for the purpose of this specification, what is considered errors may not be limited to errors on the execution units, in the relevant datapaths, or datapaths themselves, but may include errors of all types of interconnect, memory elements, clock circuitry, power distribution network and any other device that participates in data processing, storage, communication, or acquisition.
Further, in alternate embodiments, error diagnosis and/or creation of correction may be performed on another computing platform other than the computing platform performing the coordinated computation and error detection.
In various embodiments, computing platform 101 may be a wireless computing device such as a mobile phone, a media player, a laptop computer, a personal digital assistant, and so forth. Computing platform 103 on the other hand may be any one of a number of servers.
FIG. 2 illustrates a method 200 of the present disclosure to address computing errors in further detail, in accordance with various embodiments. As shown, method 200 may comprise all or a subset of the following operations: (i) Test input data generation 202; (ii) Error detection 204; (iii) Error diagnosis 206; (iv) Errors characterization 208; (v) Error corrections 210; and/or (vi) System verification 212. In block 202, in various embodiments, the goal may be to generate test vectors that cover some or all possible errors in systematic way so that more likely and the higher impact errors may be first identified. These test input may be either identical or different for different phases. At block 204, in various embodiments, error detection may be through cut-based comparison, described more fully below. At block 206, in various embodiments, the process may start with cuts that may be far from each other and gradually introduce more dense cuts in order to facilitate error diagnosis. At block 208, in various embodiments, for error characterization, the process may diagnosis the errors by tracing the variables from one cut to another, described more fully below. At block 210, in various embodiments, for error correction, a spectrum of corrections may be generated, so that user may select one that may be perceived as beneficial. In various embodiments, different corrections may be created for different quality of service (QoS) requirements of different applications. At block 212, in various embodiments, overall system verification may be conducted using e.g. learn-and-test methodology.
FIGS. 3 a-3 c illustrate the concept of cut and variable observation, and their employment for error detection, diagnosis and correction, in accordance with various embodiments of the present disclosure. More specifically, FIGS. 3 a-3 c illustrate an example execution of the first three iterations of a small program that may be executed in a very long loop. In this example small program, inputs In₁, In₂, In₃, and In₄are correspondingly multiplied with constants C₁, C₂, C₃, and C₄using multiplier M₁, M₂, M₃, and M₄respectively. The products of the first two multiplications In₁with C₁and In₂and C₂, and the products of the last two multiplications In₃with C₃and In₄and C₄are correspondingly added together using accumulators A₁and A₂respectively. The results of the additions are then further added together using accumulators A₃. The result is then outputted as well as provided as In₂for the next iteration. Further, the output of accumulators A₃is further combined with another value using accumulators A₄and provided as input In₄for the next iteration.
As illustrated, a cut at the point of the computation where outputs of the multiplications have been initially pairwise added together may be taken (i.e. at the output of accumulator A₁and A₂) in each of the three iterations. At each cut, various variables may be observed, and the observations may be used for subsequent error detection, diagnosis and correction.
Before further continuing with the present disclosure, it should be noted that any error between two cuts may be detected by checking only the variable values of the later cut. In addition, one may further reduce the average number of observed variables by checking only the cuts every k-th iterations. Further, while a cut may be taken at each of three iterations of the example execution, the present disclosure is not so restricted; in alternate embodiments, the cuts do not have to be taken periodically.
In various embodiments, the basic scheme for error detection may be by comparing the variable values observed at the cut of the less reliable (e.g. lower energy, less resource) processor or computing platform with the variable values observed at the cut of the more reliable (e.g. higher energy, more resource) processor computing platform. In various embodiments, the reliability and the accuracy of the trusted processor or computing platform may be further improved by employing high precision arithmetic, enforcement of favorable operational and environmental conditions, and double, triple or higher redundancy.
In various embodiments, the variable values of each cut may be observed. The cut variables represent a set of states and/or inputs that completely defined some or all consequent outputs. The location of a cut, and the frequency of taking cuts may be application dependent. In various embodiments, the cuts may be improved or optimized according to a variety of objectives including but not limited to their cardinality; difficulty of correcting the cuts variables, and suitability for compression and/or decompression.
In various embodiments, error diagnosis may be conducted between two consecutive cuts where some or all variables in the earlier cut may be correct. An error may be considered to be detected if at least one variable in a later cut is not correct. In various embodiments, error diagnosis further comprises backward tracing of the incorrect variables starting from the later cut. In instances where the correct value on one of the inputs is detected, the search along that direction may be terminated. Accordingly, the source for each detected error may be determined.
It should be noted that error diagnosis may not necessarily be performed for error correction, as errors may or may not have to be performed at the same places where they may be introduced. Consider the case where two chained adders have permanent manufacturing faults. The first adder, for example, increases the result by seven, and the second reduces the sum by nine systematically. It is easy to see that it may be sufficient to increase one of the inputs by two in order to always correctly rectify the two chained adders. However, the diagnosis of errors may be illuminating for design and operation of the processors or computing systems by revealing what should be altered in order to prevent a particular error or a type of error.
As alluded to earlier, in various embodiments, tracing may be performed by comparing the complete cuts of the program. The search may be conducted between the first cut where the differences between the variable values of the two computing platforms may be observed, and the previous cut. Either forward or backward search may be used. In various embodiments, the assumption may be made that there are no canceling errors, since errors that may be cancelled do not have to be addressed. Cancellation of two or more errors happens when, for example, one addition increases the output variable by the same amount, as the consequent addition reduces.
In various embodiments, error classification may be conducted by identifying the properties of errors and by grouping the actually observed errors with similar values according to one or more properties together. In various embodiments, the error properties may include the source (e.g, bitwidth, clock cycle time, operational conditions, environment), permanency (single cycle, multiple cycles, permanent), the sequential depth of combinational logic where the error may be observed, suitability for corrections according to input data, variable, and program alternations, the impact on the accuracy in terms of the impacted output variables and their change, and the percentage of devices where the pertinent error may be likely to occur. In various embodiments, error classification may be performed based at least in part on the likelihood of presence of a error type on the less reliable processor or computing platform.
In various embodiments, error correction may be conducted at the places where the errors occurred. In other embodiments, some or all errors may be corrected at the cut variables. In still other embodiments, the permanent faults may be corrected using addition program instructions inserted before or after operation executed at a faulty execution unit or stored in a faulty register.
In various embodiments, correction of errors may be conducted using any combination of mechanisms including but not limited to the following five ways: (i) data corrections; (ii) constant and variable corrections; (iii) program or computational structure corrections; (iv) operational conditions corrections; and/or (v) environment alternations. The corrections may be limited to a single data set executed using a given software and/or hardware implemented functionality, may be specific or generic with respect to the source of errors, and may target automatic correction of a specific types of errors without need to consider a specific input datasets. In various embodiments, the correction may be complete in the sense that the results on the considered platforms may be identical to one with the highest reliability and accuracy or may be partial and conducted in such a way that a specific objective error norm or subjective quality of service criteria may be satisfied. In addition, the corrections may be conducted so that a specific operational metrics such as the latency and throughput may be satisfied.
In various embodiments, error corrections may be conducted using one or more of the mechanisms that include but are not limited to program, inputs, and variables alternation and conditions and environment alternations. Selecting a way for compensation may be driven by design and operation metrics and the architecture, operating system, system software and other implementation and operational issues. For example, if communication is much more expensive than computation, than the primary goal may be to reduce the required amount of communication. Therefore, selecting a small cut where the corrections may be suitable for compression may be favored as compared to the simplicity of correcting the output or the inputs.
As described earlier, due to technological and application trends, it may be likely that it may be difficult to manufacture fault-free integrated circuits. In addition, the complexity of hardware, system and utility software may result into frequent error corrections. In order to address this, a technique for addressing error may additionally or alternatively include (i) Systematic errors correction for an arbitrary inputs to an arbitrary or a specific program/functionality; and/or (ii) Synthesis for error correction.
In various embodiments, synthesis for error correction comprises of a set of synthesis and implementation decisions that support any of the earlier described operations for systematic errors correction for an arbitrary inputs to an arbitrary or a specific program/functionality. In various embodiments, a special emphasis may be placed in error correction where hardware may be added to allow rapid correction of internal variables. In various embodiments, scheduling slots may be intentionally left empty for error correction. In other embodiments, built-in-self-repair techniques may be employed for correction of manufacturing variability errors.
FIG. 4 illustrates an example processor 400 with implementation decisions that facilitate addressing errors, including error detection and correction through coordinated computation as earlier described, in accordance with various embodiments of the present disclosure. As shown, the processor 400 includes a widely used dedicated register file architecture where some or all registers 402 may be grouped in a number of register files RF₁, RF₂, RF₃, RF₄, RF₅, and RF₆in order to share access logic and reduce interconnect, and each register file RF₁, RF₂et al may be connected to an input one of the execution units. In particular, for the illustrated embodiments, there may be three execution units arithmetic logic unit (ALU) 404, adder 406, and multiplier 408. Registers files RF1 and RF2 may be coupled to the ALU 404, register files RF3 and RF4 may be coupled to the adder 406, and register files RF5 and RF6 may be coupled to the multiplier 408. Note that the there may be direct connection from the input port only to register files RF1 and RF4 and that there may be a direct connection to the output only from the output of the multiplier. In various embodiments, in synthesis for error detection and correction, variables that are observed to be ones that are result of one of the multiplications may be stored in register files RF1 and RF4, and the corrections may be applied to variables that are stored in register files RF1 and RF4. Thus, the hardware organization making the example processor particularly suitable for practicing the earlier described approaches to error detection and/or correction.
FIG. 5 is a block diagram illustrating an example computing device configured in accordance with the present disclosure. In a very basic configuration 501, computing device 500 typically includes one or more processors 510 and system memory 520. A memory bus 530 may be used for communicating between the processor 510 and the system memory 520.
Depending on the desired configuration, processor 510 may be of any type including but not limited to a microprocessor (μP), a microcontroller (μC), a digital signal processor (DSP), or any combination thereof. Processor 510 may include one more levels of caching, such as a level one cache 511 and a level two cache 12, a processor core 513, and registers 514. An example processor core 513 may include an arithmetic logic unit (ALU), a floating point unit (FPU), a digital signal processing core (DSP Core), or any combination thereof. An example memory controller 515 may also be used with the processor 510, or in some implementations the memory controller 515 may be an internal part of the processor 510.
Depending on the desired configuration, the system memory 520 may be of any type including but not limited to volatile memory (such as RAM), non-volatile memory (such as ROM, flash memory, etc.) or any combination thereof. System memory 520 may include an operating system 521, one or more applications 522, and program data 524. Application 522 may include programming instructions providing logic to implement the above described coordinated computation based error detection and correction. Program Data 524 may include the applicable and related coordinated computation based error detection and correction data values.
Computing device 500 may have additional features or functionality, and additional interfaces to facilitate communications between the basic configuration 501 and any required devices and interfaces. For example, a bus/interface controller 540 may be used to facilitate communications between the basic configuration 501 and one or more data storage devices 550 via a storage interface bus 541. The data storage devices 550 may be removable storage devices 551, non-removable storage devices 552, or a combination thereof. Examples of removable storage and non-removable storage devices include magnetic disk devices such as flexible disk drives and hard-disk drives (HDD), optical disk drives such as compact disk (CD) drives or digital versatile disk (DVD) drives, solid state drives (SSD), and tape drives to name a few. Example computer storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, or other data.
System memory 520, removable storage 551 and non-removable storage 552 are all examples of computer storage media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which may be used to store the desired information and which may be accessed by computing device 500. Any such computer storage media may be part of device 500.
Computing device 500 may also include an interface bus 542 for facilitating communication from various interface devices (e.g., output interfaces, peripheral interfaces, and communication interfaces) to the basic configuration 501 via the bus/interface controller 540. Example output devices 560 include a graphics processing unit 561 and an audio processing unit 562, which may be configured to communicate to various external devices such as a display or speakers via one or more A/V ports 563. Example peripheral interfaces 570 include a serial interface controller 571 or a parallel interface controller 572, which may be configured to communicate with external devices such as input devices (e.g., keyboard, mouse, pen, voice input device, touch input device, etc.) or other peripheral devices (e.g., printer, scanner, etc.) via one or more I/O ports 573. An example communication device 580 includes a network controller 581, which may be arranged to facilitate communications with one or more other computing devices 590 over a network communication link via one or more communication ports 582.
The network communication link may be one example of a communication media. Communication media may typically be embodied by computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and may include any information delivery media. A “modulated data signal” may be a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media may include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), microwave, infrared (IR) and other wireless media. The term computer readable media as used herein may include both storage media and communication media.
Computing device 500 may be implemented as a portion of a small-form factor portable (or mobile) electronic device such as a cell phone, a personal data assistant (PDA), a personal media player device, a wireless web-watch device, a personal headset device, an application specific device, or a hybrid device that include any of the above functions. Computing device 500 may also be implemented as a personal computer including both laptop computer and non-laptop computer configurations.
FIG. 6 illustrates a block diagram of an example computer program product 600, all arranged in accordance with the present disclosure. In some examples, as shown in FIG. 6, computer program product 600 includes a signal bearing medium 602 that may also include programming instructions 604. Programming instructions 604 may be for receiving one or more observations of one or more variables of one or more cuts of a computation performed on another computing platform using one or more input data. Programming instructions 604 may also be for detecting one or more errors of the computation performed on the another computing platform, based at least in part on the observations of the one or more variables of the cut of the computation. Further, programming instructions 604 may be for determining a set of one or more corrections, and providing to the another computing platform, the set of one or more corrections. In some embodiments, programming instructions 604 may also be for implementing a set of provided corrections.
Also depicted in FIG. 6, in some examples, computer product 600 may include one or more of a computer readable medium 606, a recordable medium 608 and a communications medium 610. The dotted boxes around these elements depict different types of mediums included within, but not limited to, signal bearing medium 602. These types of mediums may distribute programming instructions 604 to be executed by logic. Computer readable medium 606 and recordable medium 608 may include, but are not limited to, a flexible disk, a hard disk drive (HDD), a Compact Disc (CD), a Digital Video Disk (DVD), a digital tape, a computer memory, etc. Communications medium 610 may include, but is not limited to, a digital and/or an analog communication medium (e.g., a fiber optic cable, a waveguide, a wired communication link, a wireless communication link, etc.).
Claimed subject matter is not limited in scope to the particular implementations described herein. For example, some implementations may be in hardware, such as employed to operate on a device or combination of devices, for example, whereas other implementations may be in software and/or firmware. Likewise, although claimed subject matter is not limited in scope in this respect, some implementations may include one or more articles, such as a storage medium or storage media. This storage media, such as CD-ROMs, computer disks, flash memory, or the like, for example, may have instructions stored thereon, that, when executed by a system, such as a computer system, computing platform, or other system, for example, may result in execution of a processor in accordance with claimed subject matter, such as one of the implementations previously described, for example. As one possibility, a computing platform may include one or more processing units or processors, one or more input/output devices, such as a display, a keyboard and/or a mouse, and one or more memories, such as static random access memory, dynamic random access memory, flash memory, and/or a hard drive.
There is little distinction left between hardware and software implementations of aspects of systems; the use of hardware or software is generally (but not always, in that in certain contexts the choice between hardware and software may become significant) a design choice representing cost vs. efficiency tradeoffs. There may be various vehicles by which processes and/or systems and/or other technologies described herein may be effected (e.g., hardware, software, and/or firmware), and that the favored vehicle may vary with the context in which the processes and/or systems and/or other technologies are deployed. For example, if an implementer determines that speed and accuracy are paramount, the implementer may opt for a mainly hardware and/or firmware vehicle; if flexibility is paramount, the implementer may opt for a mainly software implementation; or, yet again alternatively, the implementer may opt for some combination of hardware, software, and/or firmware.
In some embodiments, several portions of the subject matter described herein may be implemented via Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), digital signal processors (DSPs), or other integrated formats. However, those skilled in the art will recognize that some aspects of the embodiments disclosed herein, in whole or in part, may be equivalently implemented in integrated circuits, as one or more computer programs running on one or more computers (e.g., as one or more programs running on one or more computer systems), as one or more programs running on one or more processors (e.g., as one or more programs running on one or more microprocessors), as firmware, or as virtually any combination thereof, and that designing the circuitry and/or writing the code for the software and or firmware would be well within the skill of one of skill in the art in light of this disclosure. In addition, those skilled in the art will appreciate that the mechanisms of the subject matter described herein are capable of being distributed as a program product in a variety of forms, and that an illustrative embodiment of the subject matter described herein applies regardless of the particular type of signal bearing medium used to actually carry out the distribution. Examples of a signal bearing medium include, but are not limited to, the following: a recordable type medium such as a floppy disk, a hard disk drive, a Compact Disc (CD), a Digital Video Disk (DVD), a digital tape, a computer memory, etc.; and a transmission type medium such as a digital and/or an analog communication medium (e.g., a fiber optic cable, a waveguide, a wired communications link, a wireless communication link, etc.).
The herein described subject matter sometimes illustrates different components or elements contained within, or connected with, different other components or elements. It is to be understood that such depicted architectures are merely examples, and that in fact many other architectures may be implemented which achieve the same functionality. In a conceptual sense, any arrangement of components to achieve the same functionality is effectively “associated” such that the desired functionality is achieved. Hence, any two components herein combined to achieve a particular functionality may be seen as “associated with” each other such that the desired functionality is achieved, irrespective of architectures or intermedial components. Likewise, any two components so associated may also be viewed as being “operably connected”, or “operably coupled”, to each other to achieve the desired functionality, and any two components capable of being so associated may also be viewed as being “operably couplable”, to each other to achieve the desired functionality. Specific examples of operably couplable include but are not limited to physically mateable and/or physically interacting components and/or wirelessly interactable and/or wirelessly interacting components and/or logically interacting and/or logically interactable components.
It will be understood by those within the art that, in general, terms used herein, and especially in the appended claims (e.g., bodies of the appended claims) are generally intended as “open” terms (e.g., the term “including” should be interpreted as “including but not limited to,” the term “having” should be interpreted as “having at least,” the term “includes” should be interpreted as “includes but is not limited to,” etc.). It will be further understood by those within the art that if a specific number of an introduced claim recitation is intended, such an intent will be explicitly recited in the claim, and in the absence of such recitation no such intent is present. For example, as an aid to understanding, the following appended claims may contain usage of the introductory phrases “at least one” and “one or more” to introduce claim recitations. However, the use of such phrases should not be construed to imply that the introduction of a claim recitation by the indefinite articles “a” or “an” limits any particular claim containing such introduced claim recitation to inventions containing only one such recitation, even when the same claim includes the introductory phrases “one or more” or “at least one” and indefinite articles such as “a” or “an” (e.g., “a” and/or “an” should typically be interpreted to mean “at least one” or “one or more”); the same holds true for the use of definite articles used to introduce claim recitations. In addition, even if a specific number of an introduced claim recitation is explicitly recited, those skilled in the art will recognize that such recitation should typically be interpreted to mean at least the recited number (e.g., the bare recitation of “two recitations,” without other modifiers, typically means at least two recitations, or two or more recitations). Furthermore, in those instances where a convention analogous to “at least one of A, B, and C, etc.” is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., “a system having at least one of A, B, and C” would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.). In those instances where a convention analogous to “at least one of A, B, or C, etc.” is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., “a system having at least one of A, B, or C” would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.). It will be further understood by those within the art that virtually any disjunctive word and/or phrase presenting two or more alternative terms, whether in the description, claims, or drawings, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase “A or B” will be understood to include the possibilities of “A” or “B” or “A and B.”
Although certain embodiments have been illustrated and described herein for purposes of description of the preferred embodiment, it will be appreciated by those of ordinary skill in the art that a wide variety of alternate and/or equivalent embodiments or implementations calculated to achieve the same purposes may be substituted for the embodiments shown and described without departing from the scope of the disclosure. Those with skill in the art will readily appreciate that embodiments of the disclosure may be implemented in a very wide variety of ways. This application is intended to cover any adaptations or variations of the embodiments discussed herein. Therefore, it is manifestly intended that embodiments of the disclosure be limited only by the claims and the equivalents thereof.

Claims

1. A method, comprising:

performing a computation by a first computing platform in coordination with a second computing platform performing a similar or the same computation;

receiving by the first computing platform, an observation of one or more variables taken at a cut of the computation performed on the second computing platform; and

detecting by the first computing platform, one or more errors of the computation performed on the second computing platform, based at least in part on the observation of the one or more variables received, and based at least in part on one or more variables taken at a corresponding cut of the computation performed on the first computing platform.

2. The method of claim 1, further comprising instructing by the first computing platform, the second computing platform on where and when to take a cut of the computation.

3. The method of claim 1, further comprising:

determining by a third computing platform one or more corrective actions to correct the one or more detected errors; and

providing by the third computing platform, to the second computing platform, instructions about the one or more corrective actions.

4. The method of claim 3, wherein the first and third computing platforms are the same computing platforms.

5. The method of claim 3, wherein the method further comprises receiving by the first computing platform, a second observation for one or more variables at a second cut of the computation performed on the second computing platform, taken by the second computing platform at a first point in time, and the second observation of one or more variables at the second cut of the computation received by the first computing platform is taken by the second computing platform at a second later point in time, wherein the computation is k iteration later at the second later point in time, when compared to the computation at the first point in time, where k is an integer greater than or equal to 1.

6. The method of claim 5, wherein determining comprises tracing by the third computing platform, the execution between the first and second points in time.

7. The method of claim 5, wherein at least one of the detection or determination is performed in view of a likelihood of presence of a type of error at the second computing platform or in view of a characteristic of the second computing platform.

8. The method of claim 5, wherein the method further comprises receiving by the first computing platform, a third observation for one or more variables at a third cut of the computation performed on a second computing platform, taken by the second computing platform at a third point in time later than the second point in time by an amount that is different from an amount the second point in time is later than the first point in time.

9. The method of claim 3, wherein providing comprises providing one or more instructions about a corrective action to correct a detected error at a selected one of a location where the error occurred or at a variable.

10. The method of claim 3, wherein providing comprises providing one or more instructions about a corrective action to correct a detected error by inserting one or more instructions into the computation at a location before or after an instruction is executed at a faulty execution unit of the second computing platform, or at a location before or after a result of an instruction is stored into a faulty register of the second computing platform.

11. A method comprising:

performing a computation, by a first computing platform, using one or more input data, the first computing platform configured to perform a similar or the same computation as a second computing platform in a coordinated manner;

observing one or more variables at a cut of the computation, by the first computing platform;

sending the observation of the one or more variables at the cut of the computation from the first computing platform to the second computing platform; and

receiving instructions from the second computing platform by the first computing platform, about one or more corrective actions for one or more errors of the computation performed on the first computing platform.

12. The method of claim 11 wherein observing one or more variables comprises observing the one or more variables at a first point in time, and the method further comprises the first computing platform repeating the observing for another cut at a second later point in time, and then, repeating the sending.

13. The method of claim 12, wherein the computation is k iteration later at the second later point in time, when compared to the computation at the first point in time, where k is an integer greater than or equal to 1.

14. The method of claim 12, further comprising observing one or more variables of a third cut of the computation at a third point in time, by the first computing platform, wherein the third point in time is later than the second point in time by an amount that is different from an amount the second point in time is later than the first point in time.

15. The method of claim 11, further comprising receiving instructions on the one or more corrective actions from the second computing platform with the first computing platform.

16. The method of claim 11, further comprising performing the one or more corrective actions with the first computing platform.

17. The method of claim 16, wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to correct a detected error at a selected one of a location where the error occurred or at a variable.

18. The method of claim 16, wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to correct a detected error by inserting one or more instructions into the computation at a location before or after an instruction is executed at a faulty execution unit of the first computing platform, or at a location before or after a result of an instruction is stored into a faulty register of the first computing platform.

19. The method of claim 16, wherein observing comprises reading a first register of the first computing platform; and receiving comprises receiving instructions to take one or more corrective actions at a second register of the first computing platform that is different from the first register.

20. The method of claim 16 wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to correct a detected error by controlling an operating or environmental condition of the first computing platform.

21. The method of claim 16, wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to partially correct the one or more detected errors in view of one or more quality of service criteria of the first computing platform.

22. The method of claim 16, wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to correct a detected error by causing a circuit of the first computing platform to be invoked under a compilation or execution time condition.

23. The method of claim 16, wherein receiving instructions comprises receiving instructions from the second computing platform by the first computing platform, to add an application specific or reconfigurable circuit to the first computing platform to address one of the one or more detected errors.

24. The method of claim 11, further comprises operating the first computing platform with an energy consumption rate that is smaller than an operational energy consumption rate of the second computing platform.

25. An apparatus comprising:

circuitry configured to perform a computation in coordination with another apparatus performing a similar or the same computation; and

a controller coupled to the circuitry and configured to receive an observation of one or more variables at a cut of the computation performed on the other apparatus, and to detect one or more errors of the computation performed on the other apparatus, based at least in part on the one or more observations of the one or more variables at the cut of the computation performed on the other apparatus, and one or more observations of the variables at a corresponding cut of the similar or same computation performed on the apparatus.

26. The apparatus of claim 25, the controller is further configured to instruct the other apparatus on where to take a cut of the computation, and when to observe the one or more variables at the cut of the computation.

27. The apparatus of claim 25, wherein the controller is further configured to determine one or more corrective actions to correct the one or more detected errors; and provide to the other apparatus, instructions about the one or more corrective actions.

28. The apparatus of claim 25, wherein the controller is further configured to receive second observation for one or more variables at another cut of the computation performed on the other apparatus, taken by the other apparatus at a first point in time, wherein the observation of one or more variables of a cut of the computation is taken by the other apparatus at a second later point in time, and wherein the computation is k iteration later at the second later point in time, when compared to the computation at the first point in time, where k is an integer greater than or equal to 1.

29. The apparatus of claim 28, wherein the controller is further configured to trace the execution between the first and second points in time.

30. An apparatus comprising:

circuitry configured to perform a computation using one or more input data, in a coordinated manner with another apparatus performing a similar or the same computation; and

a controller coupled to the circuitry and configured to observe one or more variables at a cut of the computation,

send the observation of the one or more variables at the cut of the computation from the apparatus to the another apparatus for the another apparatus to detect one or more errors of the computation performed on the apparatus, and to determine one or more corrective actions to correct the one or more detected errors; and

receive instructions about the one or more corrective actions to correct one or more of the detected errors.

31. The apparatus of claim 30, wherein the controller is further configured to repeat the observation for another cut at a later point in time, and then, repeat the sending of the observation.

32. The apparatus of claim 31, wherein the controller is configured to repeat the observation and sending when the computation is k iteration later at the later point in time, where k is an integer greater than or equal to 1.

33. The apparatus of claim 30, wherein the apparatus further comprises first and second register files, and for the controller is configured to observe variables stored in the first register file, and apply corrective actions to variables stored in the second register file.

34. The apparatus of claim 30, wherein the apparatus is wireless mobile computing platform.

35. An article of manufacture comprising:

a computer readable storage medium; and

a plurality of programming instructions stored therein, the programming instructions configured to program and enable an apparatus to:

receive an observation of one or more variables taken at a cut of a computation performed on another apparatus;

detect for one or more errors of the computation performed on the other apparatus, based at least in part on the one or more observations of the one or more variables taken at the cut of the computation, and one or more observations of the variables at a corresponding cut of the computation performed on the apparatus;

determine one or more corrective actions to correct the one or more detected errors; and

provide to the other apparatus, instructions about the one or more corrective actions.

36. The article of claim 35 wherein the programming instructions further enable the apparatus to

perform a similar or the same computation in coordination with the another apparatus, and

take the one or more observations of the variables at the corresponding cut of the computation performed on the apparatus.

37. The article of claim 35 wherein the programming instructions further enable the apparatus to instruct the another apparatus where to make the cut of the computation, and when to take observation of the one or more variables at the cut.

38. An article of manufacture comprising:

a computer readable storage medium; and

perform a computation using one or more input data in a coordinated manner with another apparatus performing a similar or the same computation;

observe one or more variables at a cut of the computation at a first point in time;

send the observation of the one or more variables at the cut of the computation at the first point in time from the apparatus to another apparatus for the other apparatus to detect for one or more errors of the computation performed on the apparatus, and for the other apparatus to determine one or more corrective actions to correct the one or more of detected errors; and

receive instructions from the other apparatus about the one or more corrective actions to correct the one or more of detected errors.

39. The article of claim 38, wherein the programming instructions further enable the apparatus to

observe one or more variables at another cut of the computation at a second point in time; and

send the observation of the one or more variables at the other cut of the computation at the second point in time from the apparatus to the another apparatus.